An Inversion Method of Aerosol Particle Size Distribution in Guangzhou

The relationship of ambient aerosol and visibility deterioration over Pearl River Delta(PRD) have attached more and more attentions in recent years. The extinction coefficient of ambient aerosol can be calculated with the Mie theory(N. Ma,2014), which is based on a set of measured dry aerosol number size distribution, ambient relative humidity, aerosol hygroscopic growth factor, and the assumption of no activation. Using the parameters that can be easily measured would make the extinction coefficient of ambient aerosol calculation more widely available. PM2.5 (total mass concentration of dry aerosols with the aerodynamic diameter smaller than 2.5μm) measurements are widely applied in PRD, the aerosol concentrations could be estimated based on PM2.5 measurements and used to calculate the extinction. However, with different size distributions, aerosol with the same mass concentration may have different extinction coefficients. Ignoring the variations of the shapes of aerosol size distributions may introduce an uncertainty in the calculation of aerosol extinction coefficient. In order to quantify this uncertainty, the historical data of aerosol size distribution need to be analyzed. In this paper, continuous measurements of particle number size distributions and PM2.5 were simultaneously performed at Guangzhou urban site from Nov. 2014 to Jan. 2015. The temporal and diurnal statistical results of dry season in Guangzhou were used in the inversion method of aerosol particle size distribution. The uncertainty by the parameters of the method were investigated and the extinction coefficients corresponding to a certain aerosol volume concentration and relative humidity are given in the form of probability distribution.

Acknowledgements: This work is supported by the Natural Science Foundation of China(41405133), National Key Project of MOST (2016YFC0202003) and the Natural Science Foundation of Guangdong Province, China(2014A030313788). I also acknowledge Prof. Chunsheng Zhao, Dr. Haobo Tan and Dr. Xuejiao Deng for their suggestions. The sampling and data processing supported from Dr. Shenzhen zhou, Mr. Hanbin Xu, Ms. Li Liu and Mr. Mingfu Cai are greatly appreciated.